Systemic Inflammatory Response Syndrome in Nonhuman Primates Culminating in Multiple Organ Failure,Acute Lung Injury,and Disseminated Intravascular Coagulation

Abstract

The systemic inflammatory response syndrome (SIRS) is a clinicopathological manifestation of overexuberant acute-phase inflammation caused by infectious or noninfectious etiologies. The systemic release of pro-inflammatory cytokines, chemokines, and lipid and vasoactive mediators induces endothelial damage and microvascular thrombosis, potentially culminating in disseminated intravascular coagulation (DIC), acute respiratory distress syndrome (ARDS), and multiple organ dysfunction (MOD) or failure (MOF). We present five cases in the pig-tailed macaque and olive baboon where SIRS resulted in MOF, ARDS, DIC, and the Waterhouse-Friderichsen syndrome; each with gross and histological elements manifested as edema, deposition of fibrin, hemorrhage, and thrombosis. In the described cases, SIRS was the end-common pathway for multiple risk factors that parallel those documented in humans: major surgery, obstetric complications, and infection. The diagnosis of SIRS should be considered when evaluating nonhuman primate (NHP) cases of MOF manifesting with histological evidence of vascular leakage. Experimental manipulation of NHP models may be complicated by SIRS and accompanying rapid clinical decompensation. Such adverse events may compromise toxicological studies and should be avoided when possible.

The systemic inflammatory response syndrome (SIRS) describes a generalized inflammatory process resulting from numerous and varied causes (Bone et al. 1992). It is defined clinically in humans through a combination of two or more of the following conditions: hypo- (<36°C) or hyperthermia (>38°C), tachypnea (respiratory rate > 20 breaths per minute or PaCO₂ <32 mm Hg), tachycardia (heart rate > 90 beats per minute), and white blood cell counts >12,000/cu mm, <4,000/cu mm, or with >10% immature forms (left shift). Sepsis is defined in the human literature as the systemic inflammatory response to a known infection; therefore, when a bacterial infection is diagnosed (or a positive blood culture is obtained), SIRS becomes sepsis (Cavaillon and Annane 2006). The systemic inflammatory response syndrome and sepsis lead to predictable pathophysiology through similar pathogenesis, often culminating in decompensation and death. The systemic inflammatory response syndrome and sepsis are important clinical entities in humans, associated with a mortality rate ranging from 10% to 34% (Angus et al. 2001).

Described risk factors of SIRS include major abdominal injury (most commonly secondary to vehicular trauma), multiple fractures, thermal and chemical burns, pancreatitis, obstetrical surgery, bacterial translocation, sepsis, toxemia, and neoplasia (Matsuda and Hattori 2006). Although varied, each of the risk factors for SIRS has the potential to activate the innate immune system, resulting in a convergence of common sequelae. The major effector cells of the innate immune system (monocytes, macrophages, and neutrophils) are triggered through pattern recognition receptors, which activate signaling pathways and coordinate an inflammatory response. Pattern recognition receptors, such as Toll-like receptors, recognize pathogen-associated molecular patterns of infectious agents, endogenous peptides, and damage-associated molecular patterns associated with cell death (Bianchi 2007).

The innate immune system protects the host from infectious organisms in a nonspecific manner. Although such generalized responses are initially beneficial to the host, they may become deleterious when immune stimulation is excessive or combinatorial. The systemic inflammatory response syndrome represents a destructive, unbalanced immune response leading to the systemic release of pro-inflammatory cytokines, chemokines, and vasoactive amines. Circulating mediators damage endothelium and compromise vasculature, triggering the clotting cascade and potentially culminating in disseminated intravascular coagulation (DIC). Tissue injury results from the combination of an overzealous inflammatory response, vascular damage, and compromise to the microcirculation secondary to microthrombi. Although the inflammatory response is systemic, damage may manifest first or to a greater degree in individual organs. The lungs are particularly sensitive to septic or SIRS-mediated endothelial damage; vascular damage associated within the pulmonary capillaries can lead to acute lung injury (capillary leak syndrome), leakage of albumin and fibrin, alveolar flooding, and reduced oxygen exchange. Such damage presents as an acute respiratory distress syndrome (ARDS). In cases that progress to ARDS, mortality can reach 40%, resulting in up to 43,000 annual patient deaths (Frevert and Martin 2004; Rubenfeld et al. 2005). As tissue damage progresses, multiple organ dysfunction (MOD), and further, multiple organ failure (MOF) may ensue. Multiple organ failure, MOD, and ARDS are common clinical sequelae to SIRS.

In a toxicological study, infectious disease is considered an experimental accident and a serious determinant error that compromises the experiment (Keenan 1996). As adverse events, SIRS and MOF may occur in drug trials secondary to an infectious event (sepsis) or administration of an experimental compound that is an immunomodulant/stimulant. Such adverse events must be assessed to determine relatedness to the experimental compound. Occurrences of SIRS and MOF in nonhuman primates (NHP) are rarely documented in the literature; the following case report details five such adverse events, giving background as to cause and experimental manipulation. It is our intention to shed light on possible etiologies of such adverse events. This articles describes five cases of SIRS in two species of NHP (Table 1). Each case is a result of distinct experimental etiologies related to five separate experimental protocols approved by the University of Washington’s Institutional Animal Care and Use Committee. Although several different risk factors trigger the innate immune system, the resultant lesions overlap to a great extent.

Case 1

A three-month postpartum, 4.5-year-old female Macaca nemestrina underwent experimental vascular catheterization and placement of a stomach tube (transcutaneous). On two separate occasions, the gastric/epidermal seal of the gastric tube ulcerated, requiring replacement tubes (forty-five and sixty-two days prior to euthanasia). Placement of a third tube (six days prior to euthanasia) resulted in penetration of the esophagus as a result of alteration of gastric anatomy secondary to abdominal fibrous adhesions. Postoperative blood chemistry (twenty-four hours after final tube placement) demonstrated a white blood cell (WBC) count of 15,000 × 10³/μL, predominantly composed of neutrophils, 11,000 × 10³/μL. On postoperative day 6, the animal developed increasing respiratory distress. Chest radiographs demonstrated collapse of the left caudal lung lobes with pleural effusion. A urinalysis demonstrated moderate levels of urine protein and glucose; clinicopathological monitoring demonstrated hypoalbuminemia (1.8 g/dL) and an elevated γ-glutamyl transferase (GGT) (122 U/L). During clinical evaluation, the animal became hypothermic and dyspneic; euthanasia was elected. Gross lesions included dependent pitting edema and pleural, pericardial, and thoracic effusions. Effusion fluids were turbid and contained protein (500 mg/dL), free erythrocytes (250 Ery/μL), and glucose (100 mg/dL). Histologically, acute lung injury consistent with ARDS was evident (Figure 1C). Histology of the adrenal glands (Figure 1G and 1H) demonstrated frank hemorrhage and necrosis consistent with Waterhouse-Friderichesen syndrome (WHF) and sepsis (Adem et al. 2005; Cary and Kosanke 2001). Sepsis secondary to translocation of enteric flora was suspected, however, cultures were not performed in this case, as the animal had been administered antimicrobials (entrofloxacin, cephalexin, and cefazolin) postoperatively.

Case 2

A 7.7-year-old female Macaca nemestrina on a drug safety study underwent vascular catheterization, caesarian section, gastric catheter placement, jejunal resection, and hepatic lobe resection three days prior to necropsy. Clinicopathological monitoring was performed twenty-four hours post-op, and again on the day of necropsy (seventy-two hours post-op). At twenty-four hours post-op, the animal was anemic (hematocrit [HCT] of 25% with a hemoglobin [Hb] of 7.6 g/dL), hypoalbuminemic (1.4 g/dL), diffusely edematous, and demonstrated elevated liver enzymes (alkaline phosphatase [Alk] 224 U/L, serum glutamic oxaloacetic transaminase [SGOT] 164 U/L, serum glutamic pyruvic transaminase [SGPT] 123 U/L, and GGT 55 U/L); the animal was transfused with 100 mL of whole blood. Forty-eight hours later, on the day of necropsy, edema had progressed to anasarca; clinicopathological monitoring demonstrated elevated hepatic enzymes (Alk 285 U/L, SGOT 294 U/L, SGPT 246 U/L, GGT 55 U/L), azotemia (blood urea nitrogen [BUN] 6 mg/dL, and creatinine [Crt] 1.9 mg/dL), an elevated erythrocyte sedimentation rate (27 mm/h), hypoproteinemia (4.0 g/dL), hypoalbuminemia (<1.0 g/dL), and a WBC count of 13,000 × 10³/μL with neutrophils 10,400 × 10³/μL. Despite supportive veterinary care, the animal became hypothermic (94.9°F) and lethargic, with a loss of distal limb pulses and deep pain response; euthanasia was elected. Grossly, tissues were edematous; there was approximately 100 mL of ascitic fluid within the abdomen. Multiple organs, including the liver, kidneys, gastrointestinal tract, and adrenal glands, demonstrated gross hemorrhage. The lungs were atelectic. Histology confirmed multiple organ hemorrhage and necrosis involving the liver (hepatic necrosis with disassociation of hepatic chords), kidney (glomerular capillary microthrombosis consistent with DIC; Figure 1E and 1F), gastrointestinal (with concurrent vascular infarction), and adrenal glands (WHF). At necropsy, abdominal ascitic fluid, liver, spleen, and uterus were cultured; a heavy growth of Escherichia coli was isolated from all samples despite postoperative prophylactic administration of cefazolin.

Case 3

A 4.7-year-old female Macaca nemestrina underwent experimental total body irradiation and bone marrow reconstitution. Eleven months after irradiation and bone marrow engraftment, the animal was treated with 06-benzylguanine (BCNU), a chemotherapeutic agent to select for labeled engrafted cells. Despite supportive therapy, over the subsequent seventy-two days, the animal developed thrombocytopenia (decreased from 300 × 10³/μL pre-BCNU, to 3 × 10³/μL), anemia (HCT decreased from 36% pre-BCNU, to 16%, and Hb decreased from 11.1 g/dL to 5.1 g/dL), and renal failure (BUN elevation from 19 mg/dL pre-BCNU to a maximum of 74 mg/dL post-BCNU, and creatinine elevation from 0.91 mg/dL pre-BCNU to 2.0 mg/dL post-BCNU). A urinalysis performed fifty-six days post-BCNU demonstrated a large amount of occult blood, trace protein, and a specific gravity of 1.003, confirming clinical renal failure. The animal was given palliative care including blood transfusions, fluid therapy, and prophylactic antimicrobials and antifungal drugs. At seventy-two days post-BCNU treatment, euthanasia was elected. Grossly, the lungs, gastrointestinal tract, and kidneys contained multifocal regions of parenchymal hemorrhage. Histological renal glomerular vasculature microthrombosis, consistent with DIC, correlated with clinical renal failure. Acute lung injury with alveolar proteinosis and hyaline membrane formation correlated with clinical ARDS (Figure 1A and 1B). Blood and tissue cultures were not performed owing to prophylactic administration of antimicrobials and antifungal drugs.

Case 4

A 3.1-year-old male Papio cynocephalus underwent experimental aorto-iliac engraftment. Seventy-two hours post-engraftment, circulation was compromised, as evidenced by cool, immobile hind limbs, lameness, and a markedly reduced femoral pulse. The graft was surgically repaired; an aorto-iliac saddle thrombus was removed. Upon recovery from the surgical repair, the femoral pulses were noted to be weak despite fluid therapy. The following morning, approximately eighty-four hours post-experimental engraftment and approximately twelve hours post-surgical repair, the animal was found dead. Grossly, there was a mild to moderate amount of subcutaneous, mesocolic, and retroperitoneal hemorrhage. The lungs were diffusely edematous and emphysematous with lobar hemorrhage and consolidation. Histologically, there was subacute, suppurative, fibrinohemorrhagic vasculitis within the aorta and iliac arteries adjacent to the engrafted region. The right atrium contained a peracute thrombus. Acute lung injury, consistent with ARDS, was evidenced by alveolar flooding, proteinosis, and hyaline membrane formation (Figure 1D). Cultures were not obtained as the postmortem window was unknown.

Case 5

A 2.1-year-old male Papio cynocephalus underwent total body irradiation and was reconstituted forty-eight hours later with partially mismatched (mother/offspring) bone marrow. To ameliorate the potential graft immune response, graft-versus-host disease (GVHD), a CD28 antibody was administered to the host in conjunction with the bone marrow graft. Over the subsequent ten days, the animal developed elevated liver enzymes (Alk elevation from 910 U/L to 6869 U/L, GGT elevation from 39 U/L to 175 U/L, SGPT elevation from 52 U/L to 316 U/L, and total bilirubin elevation from 0.3 mg/dL to 7.8 mg/dL), azotemia (BUN elevation from 9 mg/dL to 81 mg/dL and Crt elevation from 0.5 mg/dL to 3.2 mg/dL), thrombocytopenia (decrease from 369 × 10³/μL to 1.0 × 10³/μL), hypoproteinemia (decrease from 5.4 g/dL to 3.4 g/dL), hypoalbuminemia (decrease from 2.9 g/dL to 1.2 g/dL), and leukopenia (WBC decrease from 7.3 × 10³/μL to 1.4 × 10³/μL), as evidenced through clinicopathological monitoring. The animal developed progressive edema and anasarca in combination with hepatic and renal failure. Euthanasia was elected eleven days post bone marrow administration. Grossly, the animal was cachectic with subcutaneous and tissue edema, thoracic, pericardial, and abdominal effusions. There was hemorrhage and edema of multiple organs, including the adrenal glands, pancreas, thyroid, epidermis, gastrointestinal tract, reproductive tract, central lymph nodes, and kidneys. There was histological evidence of diffuse vasculitis and multiple organ necrosis involving the liver (hepatocellular degeneration), kidney (diffuse tubular necrosis), and central nervous system (cerebral edema and hemorrhage). Additionally, acute lung injury was evidenced by alveolar flooding, hemorrhage, and accumulation of fibrin. Graft-versus-host disease was evidenced by lymphohistiocytic inflammation of the liver, vasculature, colon, and skin (in combination with epidermal necrosis). Blood and tissue cultures were not performed owing to experimental post-irradiation prophylactic administration of antibiotics and antifungal drugs.

Discussion

The innate immune response is an evolutionarily conserved first line of defense against pathogens. Conservation of genetic code across species dictates a similar host response to various stimuli. Triggering innate immunity results in clinicopathological outcomes that overlap to a great extent because of the limited repertoire of available host responses. For example, in humans and the two NHP species examined, the triggers of SIRS are varied and not directly associated with a specific outcome, yet they culminate in similar end-state pathophysiology including vascular leakage, DIC, and MOD.

Nonhuman primates represent the best animal model of many human diseases because of their close phylogenetic linkage. In the described cases, NHPs model the triggers, clinicopathology, and possible outcomes of SIRS. Case 1 demonstrates the combinatorial effect of surgical complications and trauma, which culminated in SIRS and systemic vascular leakage/edema. Placement of multiple gastric catheters likely led to introduction of bacteria into the peritoneal and thoracic cavities. Case 2 demonstrates the combinatorial effect of obstetrical surgery, abdominal surgical complications, and sepsis resulting in MOF. Hepatic necrosis occurring in this case was likely secondary to SIRS, however, the contribution of experimental compounds and repeated hepatic surgery is unknown. Case 3 demonstrates SIRS presenting as ARDS and DIC, with elective euthanasia occurring prior to overt MOF. Case 4 demonstrates the role that vascular surgery and ischemia/reperfusion injury play in the development of SIRS and acute lung injury (Sayers 2002, reviewed in Matute-Bello, Frevert, and Martin 2008). In this case, the development of acute lung injury most likely resulted from reperfusion of ischemic tissue in the lower limbs following surgical treatment to remove an aorto-iliac saddle thrombus. Ischemia/reperfusion injury is often associated with activation of endothelial cells, increased oxygen radicals, release of inflammatory mediators (e.g., tumor necrosis factor-α [TNFa]α and complement activation), leukocyte activation, and vascular leak (Carden and Granger 2000). The development of SIRS and lung injury occurs in people and in rodents following the cross-clamping of the abdominal aorta (Sayers 2002). The findings in this case suggest animals should be closely monitored for development of SIRS and MOF when ischemia-reperfusion injury is suspected.

Case 5 demonstrates SIRS and MOF resulting from a host immune response to a foreign antigen. Graft-versus-host disease is caused by donor T cells reacting to host alloantigens, which, in this case, led to severe immunosuppression. Similar to Case 1, antibiotic therapy in Cases 3 and 5 precluded cultures; however, in the latter two cases, bone marrow failure would predispose to opportunistic bacterial infections.

Hyaline membranes were not a consistent finding in the lungs of the described cases. Alveolar proteinosis, with marked accumulation of fibrin and hemorrhage, was found in each of the four cases in which pulmonary pathology was noted. The pathophysiological progression of alveolar hyaline to hyaline membranes requires time and is aided by positive pressure ventilation. As the described animals were euthanized at the onset of clinical signs, or succumbed early in disease, there may have been insufficient time for the development of “classic” hyaline membranes associated with ARDS in man.

The final outcome of an insult is determined by host immunity, health, and genetics. The immune response is to some degree a genetic phenotype that predisposes certain animals to succumb to a similar insult from which other animals may recover (Figure 2). Similar to humans (Wurfel et al. 2005), NHP populations have been shown to contain high and low immune responders; small subsets of animals will produce up to 100-fold greater cytokines (interleukins 1β, 6, 8 [IL-1β, IL-6, IL-8] and TNF-α) in response to infectious stimuli. High responders may be more susceptible to developing SIRS when exposed to a given risk factor than would an animal with a less vigorous cytokine response.

When an animal presents with clinical signs of SIRS or sepsis, rapid clinical evaluation must be made to determine the progression of disease. In advanced cases, timely euthanasia should be considered. In early cases, before advanced clinical signs, the diagnostic workup should include a blood chemistry panel (including fibrin split products), a coagulation panel, complete blood count, blood gases, blood culture, chest radiographs, blood pressure, heart rate, and respiratory rate. In cases of pulmonary distress, sedation may not be indicated, thereby reducing diagnostic tests available in animals that cannot be manipulated while awake. For research purposes a further diagnostic workup may include running a cytokine panel to determine elevation of key mediators such as IL-1β, IL-6, IL-8, and TNF-α. Further, comparison of the immune response of animals succumbing to SIRS to those surviving similar insults is necessary to delineate the impact of host genetics and immune response to development of SIRS.